Public hospitals in financial distress: Is privatization a strategic choice?

337

OctoberYDecember & 2015

Public hospitals in financial distress: Is privatization a strategic choice? Zo Ramamonjiarivelo Robert Weech-Maldonado Larry Hearld Nir Menachemi Josue´ Patien Epane´ Stephen O’Connor Background: As safety net providers, public hospitals operate in more challenging environments than private hospitals. Such environments put public hospitals at greater risk of financial distress, which may result in privatization and deterioration of the safety net. Purpose: The purpose of this study was to investigate whether financial distress is associated with privatization among public hospitals. Methodology/Approach: We used panel data merged from the American Hospital Association Annual Survey, Medicare Cost Reports, Area Resource File, and Local Area Unemployment Statistics. Our study population consisted of all U.S. nonfederal acute care public hospitals in 1997 tracked through 2009, resulting in 6,426 hospital-year observations. The dependent variable ‘‘privatization’’ was defined as conversion from public status to either private not-for-profit or private for-profit status. The main independent variable, ‘‘financial distress,’’ was based on the Altman Z-score methodology. Control variables included market and organizational factors. Two random-effects logistic regression models with state and year fixed-effects were constructed. The independent and control variables were lagged by 1 year and 2 years for Models 1 and 2, respectively. Findings: Public hospitals in financial distress had greater odds of being privatized than public hospitals not in financial distress: (OR = 4.53, p G .001) for Model 1 and (OR = 3.05, p = .001) for Model 2. Practice Implications: Privatization eases access to resources and may provide financial relief to government entities from the burden of continuously funding a hospital operating at a loss, which in turn may help keep the hospital open

Key words: financial distress, privatization, property rights theory, public hospitals, resource dependence theory Zo Ramamonjiarivelo, PhD, MBA, is Assistant Professor, Department of Health Administration, Governors State University, University Park, Illinois. Email: [email protected]. Robert Weech-Maldonado, PhD, MBA, is Professor and L.R. Jordan Endowed Chair of Health Administration, Department of Health Services Administration, University of Alabama at Birmingham. Larry Hearld, PhD, MSA, MBA, is Assistant Professor, Department of Health Services Administration, University of Alabama at Birmingham. Nir Menachemi, PhD, MPH, is Professor and Doctoral Program Director, Department of Health Care Organization and Policy, University of Alabama at Birmingham. Josue´ Patien Epane´, PhD, MBA, is Assistant Professor, Department of Health Care Administration and Policy, University of Nevada, Las Vegas. Stephen O’Connor, PhD, MBA, MPA, FACHE, is Professor, Department of Health Services Administration, University of Alabama at Birmingham. This is to confirm that no funding was received for this study. The authors have disclosed that they have no significant relationship with, or financial interest in, any commercial companies pertaining to this article. DOI: 10.1097/HMR.0000000000000032 Health Care Manage Rev, 2015, 40(4), 337Y347 Copyright B 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

338

Health Care Management Review


and preserve access to care for the community. Privatizing a financially distressed public hospital may be a better strategic alternative than closure. The Altman Z-score could be used as a managerial tool to monitor hospitals’ financial condition and take corrective actions.

F

or several decades, public or government hospitals have been operating under a challenging environment because of reductions in Medicare and Medicaid payments, diminishing funds from local or state governments, loss of Medicaid patients to private hospitals, increasing number of uninsured patients, excess capacity, and intensifying competition (Legnini et al., 1999; Sataline, 2010; Tradewell, 1998; Weil, 2011). As the ‘‘providers of last resort’’ or ‘‘safety net providers,’’ public hospitals are expected to provide health care services to everyone regardless of health insurance status or ability to pay; they have been required to provide more charity care than private hospitals (Villa & Kane, 2013). All these factors have contributed to the financial crisis that most public hospitals have been facing (Sataline, 2010). When public hospitals are in financial trouble, they become a financial burden to the government entities that own them. Privatization is defined as an ownership conversion from public status to either for-profit or not-for-profit status (Legnini et al., 1999; Tradewell, 1998). This can involve a transaction with a private entity in terms of sale of assets or merger. Privatization has been found to be one of the strategies adopted by government entities that own financially distressed hospitals (Burns, Shah, Sloan, & Powell, 2009; Legnini et al., 1999), especially when public resources are in decline (Tradewell, 1998). For instance, in May 2013, the state of Louisiana decided to cut public hospitals funding by $781 million and privatized 9 of its 10 public hospitals to reduce Medicaid spending (Bannon, 2013). On the basis of privatization literature, other factors associated with public hospital privatization are increased competition among hospitals, inability to compete for managed care and other third-party payer contracts, reductions in Medicaid reimbursement or disproportionate share hospital funding, delayed reimbursement, easier access to capital, inefficiency, and freedom from public governance constraints (Bovbjerg, Marsteller, & Ullman, 2000; Burns et al., 2009; Legnini et al., 1999; Sloan, Ostermann, & Conover, 2003; Tiemann & SchreyPgg, 2009, 2010; Weil, 2011). To date, only one study has examined the relationship between poor financial performance and ownership conversion of public and not-for-profit hospitals (Sloan et al., 2003). This study found that chronically low operating margins and high debt-to-asset ratios increased the odds of not-forprofit or public hospitals’ conversion to for-profit status. However, this study combined public hospitals and private not-for-profit hospitals into one category, limiting the conclusions that can be made with respect to privatization of public hospitals.

This study builds on prior literature by exploring the role of financial distress in public hospitals’ privatization, using a comprehensive measure of financial distress, the Altman (1968) Z-score model. This model has been widely used in the finance and health care literatures to predict financial distress (Almwajeh, 2004; Langabeer, 2006, 2008; Ramamonjiarivelo, Weech-Maldonado, Hearld, & Pradhan, 2014), and it is calculated based on a weighted composite score of four financial ratios: liquidity, financial leverage, profitability, and capital structure ratios. As such, it takes into account ‘‘the interrelationships of many different financial aspects, similar to that of a balanced scorecard’’ (Langabeer, 2006, p. 87).

Conceptual Framework This study is premised on the resource dependence theory (RDT) and property rights theory (PRT). RDT posits that ‘‘the key to organizational survival is the ability to acquire and maintain resources’’ (Pfeffer & Salancik, 1978, p. 2). Resources are the inputs that organizations need to produce products or services, and the environment represents the ‘‘organization’s source of inputs and sink of outputs’’ (Pennings & Tripathi, 1978, p. 172). As such, the organization’s environment consists of other entities from which it procures resources and to which it sells products and services. Because an organization’s possession and control of key resources imply power, organizations have to adopt various strategies to acquire and control these resources to reduce their dependence and increase their power. Mergers and acquisitions, vertical or horizontal integrations, establishment of interorganizational coalitions, and diversifications are among the various strategic moves that organizations adopt to reduce dependence on and increase control over resources (Pfeffer & Salancik, 1978). Given government ownership, public hospitals have less freedom in acquiring key financial resources. They mostly rely on government grants, revenue over costs from previous years, and the sale of tax-exempt bonds. Public hospitals’ dependence on government funds makes them financially vulnerable, because the availability of funds is contingent on economic conditions. In addition, the financial dependence of public hospitals on the government entity that owns them can impose a financial burden on the entity when hospitals operate at a loss and need to be bailed out. Unlike public hospitals, private hospitals have more flexibility in securing key financial resources. Private not-for-profit hospitals are allowed to raise capital from tax-deductible contributions


339

Privatization of Public Hospitals

of philanthropists, tax-exempt bonds, excess revenue over costs, and governmental grants. Private for-profit hospitals can raise capital by selling stocks to investors, issuing bonds, and reinvesting retained earnings (Gapenski, 2005). RDT argues that organizations exhibiting poor financial performance engage in strategic repositioning that will enhance their access to key resources. Privatization may be such a strategy. First, privatization can release the government from the burden of financially supporting a public organization operating at a loss. Prior studies have shown financial difficulty as one of the motives for public hospital privatization (Legnini et al., 1999; Sloan et al., 2003). Second, privatization can provide public organizations the freedom and access to key resources such as financial capital, updated technology, and knowledgeable human resources that may efficiently manage and boost their competitive position (Desai, Lukas, & Young, 2000) and potentially improve health care quality (Bovbjerg et al., 2000; Tiemann & Schreyo¨gg, 2009, 2010). Third, privatization releases public hospitals from the grip of government politics and bureaucracy, and it offers hospitals more freedom and flexibility in decision-making (Bovbjerg et al., 2000). Finally, privatization can be a source of financial resources for the government in terms of the proceeds from privatization, as well as a source of additional tax revenue if the organization adopts a forprofit status (Shen, 2003). Property rights is defined as the right to own an asset, the right to sell the asset, the right to earn income from the asset, the right to residual decision regarding the use of an asset, and the right to residual income from the use of the asset (Hart, 1995; Mahoney, 2005; Preker & Harding 2003). The right to residual decision consists of the right to make a decision regarding an asset, that a contract did not clearly assign to a specific person or entity, whereas the right to residual income consists of right to own what is left from net income after all debts are paid off (Tiemann & Schreyo¨gg, 2009). PRT posits that if an individual is given the right to make residual decisions and the right to earn residual income, that individual will make efficient decisions that will maximize the firm’s profit and consequently his or her residual income. Therefore, giving managers both the right to make residual decisions on asset use as well as the right to earn residual income is a ‘‘high-powered ownership incentive’’ toward efficiency and wealth maximization (Preker & Harding, 2003; Tiemann & Schreyo¨gg, 2009). All these explain why forprofit firms, which are profit maximizers, give managers full authority over managerial decisions as well as financial incentives that are positively correlated with the financial performance of the firm. In addition, based on PRT, shareholders have the right to sell their shares to other firms if they are not satisfied with the return of their investments, which acts as an additional incentive for managers to maximize shareholders’ wealth (Tiemann & Schreyo¨gg, 2009, 2010). Thus, based on PRT, for-profit ownership is the most efficient among other ownership types (Tiemann & Schreyo¨gg, 2010).

Although private not-for-profit hospitals are not profit maximizers, they are utility maximizers and their managers, boards, and trustees do not have the right to residual income (Amirkhanyan, 2007; Tiemann & Schreyo¨gg, 2010). Therefore, not-for-profit hospital managers do not have the same incentives as for-profit managers to maximize the financial return of the hospital (Preker & Harding, 2003). However, compared to public hospital managers, not-for-profit managers have greater influence on the residual decisions on asset use or on how the hospital reinvests its residual income. These decisions can result in increased organizational size and, as a result, increased managerial power. Therefore, not-for-profit hospital managers have greater incentives than public hospital managers to lower costs and increase efficiency and, as result, maximize residual income. Therefore, we would expect governments that own financially distressed hospitals to privatize these hospitals as a strategy to improve their financial performance and avert hospital closure. In summary, based on RDT and PRT tenets, we would expect financially distressed public hospitals to be more likely to privatize, and we hypothesize that: Compared to public hospitals in good financial condition, public hospitals in financial distress are more likely to be privatized by the government entities that own them.

Methods Data This study used four data sources: (a) the American Hospital Association (AHA) Annual Survey, (b) the Bureau of Health Profession’s Area Resource File, (c) the Medicare Cost Report from the Centers for Medicare and Medicaid Services, and (d) the Local Area Unemployment Statistics from the Bureau of Labor Statistics. The AHA data file consists of hospital profile variables such as ownership status, number of hospital beds, teaching status, multihospital system affiliation, and information with respect to the number of clinical and nonclinical staff. The Area Resource File data file contains demographic and economic information on counties. The Medicare Cost Report data file contains financial data; it is the most validated and widely accepted data for hospital financial analysis (Pink et al., 2005). The Local Area Unemployment Statistics data file contains estimates of monthly and annual averages of total employment, total unemployment, and unemployment rates at various geographical levels including metropolitan areas and census regions. This study used a national sample of government-owned, nonfederal, acute care, general and surgical hospitals in the United States as of 1997. These hospitals were followed from year to year until 2009. Our original sample consisted of 1,214 public hospitals. To derive the analytic sample, several exclusion criteria were applied. First, hospitals that


340


converted into skilled nursing facility (n = 6), an ambulatory care facility (n = 1), or a critical access hospital (n = 520) as well as public hospitals that closed (n = 20) and those that experienced multiple ownership conversions (n = 26) during the study period were excluded. Critical access hospitals were excluded because they have a different working environment including special Medicare reimbursement method. Second, we excluded hospitals that were acquired or merged (n = 9) during the study period. Third, hospitals without complete financial reports during any year of the study period were excluded from the sample (n = 44). Therefore, the final analytic sample consisted of a total number of hospitals ranging from 588 hospitals (9.15%) in 1997 to 395 hospitals (6.15%) in 2009 and a total of 6,426 observationyears with an average of 494 hospitals per year.

Variables Dependent and independent variables. The dependent variable was privatization coded as a dichotomous variable (‘‘1’’ = hospital privatized; ‘‘0’’ = hospital not privatized). The main independent variable was the measure of financial distress coded as a dichotomous variable (‘‘1’’ = hospital in financial distress; ‘‘0’’ = hospital not in financial distress). Financial distress is an indicator of a hospital’s financial health and its capacity to meet its debt obligations. The Altman Z-score model was used to determine whether a public hospital is in financial distress (Altman, 2002; Langabeer, 2006). The Altman Z-score model is a discriminant function derived from a multiple discriminant analysis. The multiple discriminant analysis generates a weighted linear function of the financial ratios that ‘‘best’’ discriminates between the group of firms in financial distress and those not in financial distresses (Altman, 1968, p. 592). The Altman Z-score model (Altman, 1993) has been validated for use among service and retail firms and has been adapted for use amongst hospitals (Almwajeh, 2004; Langabeer, 2006, 2008; Ramamonjiarivelo et al., 2014). The discriminant function is formulated as follows: Z = 6.56X1 + 3.26X2 + 6.72X3 + 1.05X4, where X1 = net working capital/total assets, X2 = net assets/total assets, X3 = excess revenue over expenses/total assets, X4 = fund balance/total liabilities. X1 is a measure of liquidity and consists of the ratio of net working capital to total assets, where net working capital is defined as the difference between current assets and current liabilities (Altman, 1968). X2 is the ratio of net assets (total assets j total liabilities) to total assets and is a measure of financial leverage. X3 is the ratio of excess revenue over expenses to total assets, a measure of both profitability and productivity of the firm’s assets (Altman, 1968). X4 is the ratio of book value of equity to the book value of debt, and it represents the capital structure of the hospital. It is the


extent to which the assets are financed by debt and equity. Public hospitals do not have equity per se, but the book value of equity or fund balance can be derived by subtracting total liabilities from total assets. Following Altman’s (2002) and Langabeer’s (2006) approaches, we classified a hospital as being in financial distress if its Z-score was less than 1.1; the hospital is not in financial distress if its Z-score was greater than 1.1. Control variables. This study controlled for organizational and market characteristics that may influence a public hospital’s likelihood of being privatized. Organizational characteristics included hospital size, teaching status, outpatient mix, occupancy rate, payer mix, multihospital system membership, contract management, and participation in health networks. Hospital size was measured as the total number of hospital beds. Larger organizations have more advantages than smaller ones; they can accumulate slack resources (Sharfman, Wolf, Chase, & Tansik, 1988) and achieve economies of scale (Hall & Weiss, 1967). Teaching hospitals are usually among the largest employers in the community, making them a valuable asset for the government that owns them (AHA, 2009). In addition, their size and prestige may facilitate the acquisition of resources more easily in terms of patients and endowed funds. Teaching status was measured as a dichotomous variable coded ‘‘1’’ for teaching hospitals and ‘‘0’’ for nonteaching hospitals. A hospital was considered a teaching hospital if it exhibited one or more of the following: (a) membership of the Council of Teaching Hospitals and Health Systems, (b) affiliation with a medical school, and (c) provision of residency programs. Outpatient mix was defined as the inpatient days equivalent for outpatient visits divided by total patient days. Inpatient days equivalent for outpatient visits was obtained by dividing total outpatient visits by three (Vujicic, Addai, & Bosomprah, 2009), and total patient days was obtained by summing inpatient days equivalent for outpatient visits and total inpatient days. A larger outpatient mix indicates the hospital’s ability to acquire resources and expand its market share, which can be seen as a competitive advantage. Occupancy rate was measured as total inpatient days divided by the product of total number of beds and 365 days. Occupancy rate measures the level of inpatient services utilization, efficiency, and market share. It also indicates the hospital’s ability to compete for resources. Payer mix measures the proportions of Medicare and Medicaid inpatient days. For public hospitals, payments from Medicare and Medicaid represent an important source of revenue, especially if the hospital serves a larger proportion of uninsured patients. Three interorganizational variables (membership of a multihospital system, operation under contract management, and participation in a health care network) were also included as control variables. Organizations engage in such kind of relationships to facilitate access to resources, which can act as a barrier to privatization. AHA defined a multihospital


341


system as ‘‘two or more hospitals owned, leased, sponsored, or contract managed by the central organization’’ (AHA, 2012, p. 2). These variables were dichotomous (‘‘1’’ = hospital engaged in such relationship; ‘‘0’’ = hospital did not engage in such relationship). Market characteristics were measured at the county level using the following variables: per capita income, percentage of people who were 65 years and older, number of active physicians per 1,000 persons, unemployment rate, yearly change in unemployment rate, Medicare HMO penetration, HerfindahlYHirschman Index (HHI), excess capacity, and metropolitan location. County per capita income, percentage of people who were 65 years and older, number of active physicians per 1,000 persons, and unemployment rate indicate the munificence of the environment that may result in greater demand and supply of health care services. Prior studies found percentage of people 65 years or older, per capita income, and unemployment rate to be associated with privatization (Sloan et al., 2003). Per capita income was measured as the total annual personal income in the county divided by the total number of residents in the county. The percentage of people who were 65 years and older referred to the proportion of county residents in that age group. The number of active physicians per 1,000 population was measured as the number of nonfederal practicing physicians per 1,000 county residents. Unemployment rate was defined as the percentage of unemployed labor force in the county. The yearly change in unemployment rate indicates the level of environmental dynamism; it measures the degree of resource fluctuation in the county (Ramamonjiarivelo et al., 2014). Yearly change in unemployment rate was measured as the difference between two consecutive years’ unemployment rates divided by the unemployment rate of the earlier year. Medicare HMO penetration, HHI, excess capacity, and metropolitan location were used to measure market competition. Sloan et al. (2003) found that a higher percentage of HMO enrollees were associated with higher probability of privatization. Medicare HMO penetration was measured as the number of Medicare HMO enrollees relative to the total number of Medicare eligibles in the county. HHI has been frequently used to measure market competition (Kim, 2010; Sloan et al., 2003; Weech-Maldonado, Qaseem, & Mkanta, 2009). Sloan et al. (2003) found that higher competition was associated with privatization. HHI was defined as the sum of squared market shares (acute-care patient days for individual hospital/total acute-care patient days of all the hospitals in the county) of hospitals in the county (Kim, 2010). HHI scores range between 0 and 1, where a score closer to ‘‘0’’ indicates perfect competition whereas ‘‘1’’ indicates a monopoly. When there is excess capacity, hospitals fiercely compete among themselves to attract more patients to fill their empty beds. Excess capacity was measured in terms of the average number of unoccupied beds in the county (Weech-Maldonado et al., 2009). Furthermore, metropolitan hospitals face a larger number of competitors com-

pared to nonmetropolitan hospitals (Alexander, D’Aunno, & Succi, 1996). Metropolitan location was determined according to the RuralYUrban Continuum Codes of 2003, which classified each county in the United States into metropolitan (metro) counties and nonmetropolitan (nonmetro) counties. Metropolitan location was a dichotomous variable (‘‘1’’ = metropolitan; ‘‘0’’ = nonmetropolitan).

Analysis We conducted a Pearson’s correlation analysis to assess multicollinearity problems, and Pearson’s chi-square tests as well as independent samples t tests to describe the associations of privatization with the independent and control variables. We also ran two random-effects (RE) logistic regression models with state and year fixed-effects to test our hypothesis. State fixed-effects were used to control for the differences in interstate regulatory contexts; year fixed-effects were used to control for the effect of time in panel data (Weech-Maldonado et al., 2012). The first model included independent and control variables lagged by 1 year, and the second model contained independent and control variables lagged by 2 years. We chose to use different lag periods to account for the possibility that public hospital stakeholders may not immediately decide to privatize the hospital in response to financial distress. RE models can be used to account for unobserved characteristics of the hospitals that do not vary over time and which are assumed to be uncorrelated with each of the independent variables (Wooldridge, 2006). In addition, RE models are appropriate for panel data, which are considered to have a multilevel structure because the repeated measurements of the same observations over time are clustered (Li, Lingsma, Steyerberg, & Lesaffre, 2011). The RE models in this study can be summarized as follows: Odds of privatizationt = f (financial distresst j 1, control variablest j 1) Odds of privatizationt = f (financial distresst j 2, control variablest j 2). The analyses were conducted using SAS 9.2 and xtlogit function in STATA IC/11.

Findings Correlation analysis of the independent variables did not show any correlation coefficient above .80, a typical threshold to assess multicollinearity (Field, 2009). Pearson’s chisquare tests (Table 1) showed that, over the study period, 147 public hospitals were privatized. Among privatized hospitals, 31% were in financial distress, 52% were members of multihospital systems, 31% participated in health care network, 9% were teaching hospitals, and 41% were located in metropolitan areas. The independent samples t tests (Table 2) indicated that, compared to hospitals that remained public, privatized


342



Table 1

Pearson’s chi-square tests of independent and control variables by dependent variable ‘‘privatization’’ Variable Financial distress System membership Contract management Health network Teaching status Metro vs. nonmetro location

a

No distress Distress Non-member Member No contract Contract No network Network Teaching No teaching Non-metro Metro

No privatization (%), n = 6,279a

Privatization (%), n = 147a

Pearson’s chi-square statistic

5,658 621 4,483 1,796 5,272 1,007 4,741 1,538 1,436 4,843 3,123 3,149

101 (68.71) 46 (31.29) 71 (48.30) 76 (51.70) 122 (82.99) 25 (17.01) 101 (68.71) 46 (31.29) 13 (8.84) 134 (91.16) 86 (58.90) 60 (41.10)

70.73***

(90.11) (9.89) (71.40) (28.60) (83.96) (16.04) (75.51) (24.49) (22.87) (77.13) (49.79) (50.21)

37.11*** 0.10 3.57* 16.18*** 4.74**

n represents observation-years.

*p e .10. **p e .05. ***p e .0001.

hospitals tended to be smaller and had lower occupancy rate, higher percentage of Medicare inpatient days, and lower percentage of Medicaid inpatient days. In addition,

privatized hospitals tended to be located in counties with lower per capita income, higher percentage of population of 65 years and older, and lower Medicare HMO penetration.

Table 2

Independent samples t tests of control variables by dependent variable privatization

Variable Hospital beds Occupancy rate Outpatient mix Percentage of Medicare inpatient days Percentage of Medicaid inpatient days Per capita income Unemployment rate Percentage of population Q 65 Physicians per 1,000 population Excess capacity HerfindahlYHirschman Index Medicare HMO penetration Change in unemployment rate a

No privatization (n = 6,279)

Privatization (n = 147)

na

na

Mean

SDb

Mean

t

6,279

181.94

201.01

146

118.79

148.37

3.77***

6,189 6,279 6,279 6,279 6,272 6,272 6,264 6,264 6,266 5,683 6,272 6,254

0.56 0.43 0.45 0.24 26,437.68 5.73 13.64 1.86 57.05 0.85 9.39 0.05

0.19 0.25 0.20 0.18 8,840.71 2.76 3.67 2.02 37.16 0.31 13.67 0.24

145 146 146 146 146 146 146 146 145 129 146 147

0.50 0.41 0.49 0.20 24,502.84 5.40 14.46 1.62 46.12 0.92 6.71 0.01

0.22 0.29 0.23 0.19 7,516.14 2.70 3.81 1.43 32.44 0.23 11.90 0.19

3.94*** 1.12 -2.54* 2.70** 2.60** 1.42 -2.66** 1.43 3.51*** -2.81** 2.34* 2.14 *


b

SDb

Standard deviation.

*p e .05. **p e .01. ***p e .001.


343


In addition, privatized hospitals tended to be located in relatively less competitive markets in terms of excess capacity and HHI and in more dynamic environment in terms of yearly change in unemployment rate. The independent samples t tests that examined the factors associated with privatization to not-for-profit and forprofit status (table not shown) indicated that, among the 147 hospitals that were privatized, 80% (n = 117) were privatized into not-for-profit status. Compared to public hospitals that were privatized to not-for profit status, public hospitals that were privatized to for-profit status were on average smaller (98 beds vs. 124 beds) and had lower occupancy rate (46% vs. 51%). In addition, public hospitals that were privatized to for-profit status tended to be located in counties with higher

unemployment rate (6.48% vs. 5.14%), higher competition in terms of excess capacity (55 beds vs. 44 beds) and HHI (0.82 vs. 0.94), lower Medicare HMO penetration (5% vs. 7%), and greater yearly change in unemployment rate (0.04 vs. 0.002) compared with public hospitals that were privatized to not-for-profit status. Table 3 summarizes the results of the two RE logistic regression models; both models supported our hypothesis. Financial distress was significantly associated with privatization. The odds of privatization for hospitals in financial distress were 4.5 times greater than the odds of privatization for hospitals not in financial distress for Model 1 (p G .001); the odds were 3.1 times greater for Model 2 (p = .001). With respect to organizational factors, both models found that, compared

Table 3

Results of random-effects logistic regressions Model 1a (n = 5,123b)

Model 2c (n = 4,561b)

95% CId

95% CId

Privatization

OR

SE

Lower

Upper

OR

SE

Lower

Upper

Independent variable: Financial distress Control variables: Organizational factors Hospital beds Teaching status System membership Contract management Health network Occupancy rate Outpatient mix Percentage of Medicare inpatient days Percentage of Medicaid inpatient days Control variables: Market factors Per capita income Unemployment rate Percentage of Q65 Physicians/1,000 pop HerfindahlYHirschman Index Excess capacity Medicare HMO penetration Metro vs. nonmetro location Change in unemployment rate Overall Wald chi-square test

4.45***

1.22

2.60

7.62

3.05***

1.05

1.55

6.00

1.00 0.34* 2.53*** 0.66 0.87 0.24* 0.73 0.33 0.73

0.002 0.20 0.67 0.20 0.21 0.19 0.55 0.25 0.59

1.00 0.11 1.51 0.37 0.54 0.05 0.17 0.07 0.15

1.002 1.10 4.24 1.18 1.41 1.11 3.17 1.47 3.50

1.00 0.16** 1.88** 0.96 0.82 0.55 0.96 0.67 1.09

0.002 0.13 0.59 0.33 0.22 0.49 0.81 0.61 1.01

0.99 0.04 1.01 0.49 0.49 0.10 0.18 0.11 0.18

1.00 0.73 3.47 1.86 1.38 3.18 5.01 3.95 6.71

1.06** 0.98 1.05* 0.95 1.42 0.99 0.99 1.41 0.81

0.03 1.01 0.05 0.88 0.03 0.99 0.14 0.71 0.86 0.43 0.01 0.98 0.01 0.96 0.38 0.83 0.53 0.22 183.05***

1.12 1.08 1.12 1.26 4.68 1.00 1.02 2.39 2.91

1.04 1.02 1.01 0.93 0.41 1.00 0.99 1.32 1.20

0.04 0.98 0.05 0.92 0.04 0.94 0.16 0.67 0.25 0.12 0.01 0.99 0.02 0.96 0.39 0.73 0.77 0.34 134.34***

1.12 1.13 1.09 1.29 1.33 1.01 1.02 2.37 4.22

Year dummy and state code dummy variables were included in the analysis. a

Model with independent and control variables lagged by 1 year.

b


c

Model with independent and control variables lagged by 2 years.

d

CI means confidence interval.

*p e .1. **

p e .05.

***

p e .001.


344


to stand-alone public hospitals, the odds of privatization for public hospitals that were members of multihospital systems were 2.5 times greater (p G .001) for Model 1 and 1.9 times greater (p = .05) for Model 2. In addition, teaching status was marginally associated with lower odds of privatization compared to nonteaching status (OR = 0.34, p = .07) for Model 1, and it was significantly associated with privatization for Model 2 (OR = 0.16, p = .02). This suggests that, relative to a nonteaching hospital, a teaching hospital is less likely to be privatized in the following year and even less likely to be privatized 2 years later. Finally, hospitals with higher occupancy rates had marginally significant lower odds of privatization in Model 1 (OR = 0.24, p = .07). However, occupancy rate was not significant for Model 2, indicating that higher occupancy rate, during the year prior to privatization, has a marginal impact on privatization. We also found that market factors per capita income and percentage of the population of 65 years and older were positively associated with privatization for Model 1 (OR = 1.06, p = .03 and OR = 1.05, p = .09, respectively). However, these market factors were not significantly associated with privatization in Model 2. This suggests that these market factors are more strongly related to privatization the year before, but not 2 years before privatization. None of the market factors were significantly associated with privatization for Model 2, indicating that the external environment, 2 years prior to privatization, has no significant impact on public hospital privatization.

Discussion and Practice Implications The purpose of this study was to explore whether financial distress precedes privatization. Our study builds upon and extends previous research in this area by applying the Altman Z-score to predict privatization. Our findings suggest that public hospitals in financial distress are more likely to be privatized compared with those not in financial distress. Our finding is consistent with Sloan et al. (2003); they found that public hospitals included in the same category as not-for-profit hospitals, with chronically low operating margin and high debt-to-asset ratio, were more likely to be privatized. Converting a financially distressed hospital into a private entity could be beneficial in attracting financial capital, which is needed to redress the hospital’s financial situation. Privatization of financially distressed hospitals may also provide some relief to local governments from the financial drain of a hospital continuously operating at a loss. Moreover, keeping the hospital open through privatization helps the government preserve access to care for its residents and prevent major job loss, which can be detrimental to the local economy. This study also found that teaching public hospitals are less likely to be privatized compared to nonteaching public hospitals. Teaching hospitals are compensated for the costs


of their graduate medical education (GME) activities. The federal government is the largest single payer of GME, which contributes up to $9.5 billion per year from Medicare funds and $2 billion from Medicaid funds (Health Affairs, 2012). GME also receives some state funding. For example, states donated a total of roughly $3.78 billion to GME through their Medicaid program (Health Affairs, 2012). Given that the federal and state governments have a stake in teaching hospitals and teaching hospitals have some additional regulations than nonteaching hospitals, the decision to privatize a teaching hospital may include some highly influential and heterogeneous stakeholders (Shuit, 1996; Tradewell, 1998). PRT suggests that, as the number of stakeholders increases and as they become more diverse, the probability of reaching a consensus on property rights decision decreases (Mahoney, 2005). Also, stakeholders may not be willing to privatize teaching hospitals given their considerable contribution to the economy in terms of jobs and additional businesses created to meet the operating needs of a teaching hospital (AHA, 2009). Privatization may implement radical restructuration of the teaching hospital resulting in job losses (Shuit, 1996). In addition, because teaching hospitals are larger than their nonteaching counterparts, they may be resistant to change because of their structural inertia (Hannan & Freeman, 1984). The study also found that public hospitals that were members of multihospital systems are more likely to be privatized than stand-alone hospitals. The financial burden that a struggling hospital member imposes on the whole system may put pressure on the government to privatize the hospital. Prior studies suggest that hospitals in financial crisis seek affiliation with multihospital systems to enhance their financial performance (McCue & Furst, 1986). Therefore, if system affiliation does not lead to a better financial condition, then privatization could be the second alternative. In addition, privatizing a public hospital member of a multihospital system may not dramatically impact access to care for the indigent population who can seek care in other public hospital members of the system. Thus, privatization might have a positive impact on the remaining public hospital members of the system in terms of increased occupancy rate and increased number of outpatient visits. In some cases, the government that owns a multihospital system may decide to privatize the whole system, which consequently increases the odds of privatization of hospital members of a multihospital system (Legnini et al., 1999). Public hospitals with higher occupancy rate are less likely to be privatized; a finding consistent with that of Amirkhanyan (2007), who investigated the privatization of public nursing homes. A higher occupancy rate suggests that the public hospital is in a stronger strategic position to attract more patients than competitors. Sloan et al. (2003) suggest that higher occupancy rate indicates that the hospital is able to generate greater inpatient revenue. Greater inpatient revenue


345


may lead to higher financial performance, which decreases the odds of privatization (Sloan et al., 2003). Our findings also suggest that a more munificent environment in terms of higher per capita income and higher percentage of population of 65 years and older are associated with greater odds of privatization. On the one hand, our findings may seem surprising because one might assume that these environments would contribute to better financial performance. On the other hand, when people have higher income and health insurance, they tend to go to private hospitals instead of public hospitals. Therefore, public hospitals might be privatized to attract people with higher income and those enrolled in Medicare HMOs. With respect to market factors, only per capita income and percentage of population of 65 years and older are associated with privatization for Model 1, and none of the market factors predict privatization for Model 2. This lack of substantial findings is inconsistent with RDT that suggests that competing for resources explains organization’s strategic behavior. This finding may be because of the fact that different entities own public hospitals. Unlike private hospitals, public hospitals can be owned by a state, a city, or a county. However, the environmental variables are measured at county level. Therefore, even if the county in which the hospital operates is less munificent, more dynamic, or more complex, these situations may not directly affect the public hospital if the state or the city that owns that hospital is in good financial condition and if the state possesses adequate funding policies to keep its mission in serving the poor. In the same vein, the decision to privatize public hospitals may be because of budget shortfalls that a state or local government experiences. Health care usually represents a large proportion of government budget (Bachrach, Braslow, & Karl, 2012). Therefore, privatizing public hospitals may free up a substantial amount of funds that the government can use to finance other public projects (Bannon, 2013). Governments finance several portfolios such as health care, education, public safety and security, libraries, museums, and infrastructures, among others that compete for scarce resources. Future research should explore the role of government’s finances on public hospital privatization. Our findings also suggest that there may be noneconomic factors that may influence the decision to privatize public hospitals. From its inception, privatization has involved political forces. Therefore, privatization may be studied from the government’s perspective. Indeed, public choice theory suggests that political leaders impose their own agendas to please voters and remain in power (Tiemann & Shreyo¨gg, 2009, 2010).

Managerial and Policy Implications The findings from this study have a number of managerial and policy implications. Privatization was found to be a strategy adopted by governments that own financially dis-

tressed public hospitals. Public hospitals have played an important role in health care delivery, yet they are highly resource intensive. Privatization is probably a better alternative than closure. Keeping the hospital open under private ownership may preserve access to care at least for some vulnerable populations. For instance, it has been argued that private notfor-profit hospitals oftentimes act as safety net providers. Therefore, privatizing public hospitals to not-for-profit hospitals may not alter the provision of uncompensated care (Bovbjerg et al., 2000; Desai et al., 2000). However, conversion of public hospitals into for-profit status has raised concerns that converted hospitals may shut down expensive services and might eventually close the hospital if it does not exhibit good financial performance (Sataline, 2010). Although this study focused on the association between financial distress and privatization, this finding does not imply that privatization is the only strategy that governments that own financially distressed hospitals can adopt. Other strategies include bankruptcy filing or closure (Bazzoli & Cleverly, 1994), diversification (Alexander et al., 1996), cuts on services or asking for bail-out from the federal government (California Association of Public Hospitals and Health Systems, 2003), among others. Therefore, government constituencies may evaluate different alternatives before making a decision on how to address a hospital’s financial distress. Also, this finding does not suggest that financial distress is the only reason for privatization; other reasons include change in the mission of the government, need for politicians to please voters, difficulty in raising capital, cumbersome procurement process and inefficient compensation systems, as well as mandate to craft competitive strategies in open public meetings (Legnini et al., 1999; Tiemann & Shreyo¨gg, 2009, 2010). In addition, our findings showing that certain types of hospitals (e.g., multisystem hospitals, nonteaching hospitals, low-occupancy hospitals) are more likely to privatize may help practitioners and policy makers monitor trends that contribute to privatization, above and beyond financial distress. The Altman Z-score, which is a managerial tool to monitor the financial condition of an organization, has not been widely applied in health care management and health services research. Although the primary purpose of using the Altman Z-score is to detect financial distress, it could be used as a managerial tool to regularly assess the financial condition of the organization (Calandro, 2007) and consequently address the problem before it is too late. Finally, given the major role of public hospitals as safety net providers, there are concerns that privatization may reduce access to care for the indigent as converted hospitals might not be as committed to serving the poor (Desai et al., 2000). Privatization may results in price increase, lower quality of care, and reduced community benefits (Shen, 2003). However, to the extent that financially distressed public hospitals are privatized, this may ensure continued operations for a hospital that may eventually close.


346


Study Limitations and Conclusion There are several limitations to this study. First, because of the small number of public hospital closures (n = 20), we were not able to recalibrate the Altman Z-score for use in this study. The validation of the Altman Z-score for use in public hospital studies is an area for future research. Second, because of data limitations, this study did not include measures of the political environment. Because public hospitals’ operating environments are highly politicized and the government that owns the hospital is the ultimate decisionmaker on privatization, controlling for political environment might help explain privatization. Third, because of lack of data on the financial situation of the governments that own public hospitals, which may be a factor leading to privatization, we were not able to explain the role of government financial situation in privatization. Fourth, we included multihospital system membership as a variable in the study. Given that the individual hospital is the unit of analysis and because of data limitation, we are not able to explain the role of the system as a whole on privatization, in terms of its financial situation and number of hospitals in the system. Fifth, another limitation of this study was the lack of data with respect to the proportion of privately insured, underinsured, and uninsured patients. Including the proportion of privately insured and uninsured patients could have given additional insights on whether serving patients other than those enrolled in Medicare and Medicaid has an impact on privatization. Despite these limitations, this study provides important insights into the role of financial distress and other organizational and market factors on governments’ decisions to privatize public hospitals. Additional empirical studies that include political and government financial factors to predict public hospitals’ privatization are needed to fill the gap in this area of study. Future studies are also needed to fill the gap on the antecedents of the privatization of multihospital systems and teaching hospitals. Finally, based on our definition of privatization, corporatization of public hospitals was beyond the scope of this study. The determinants of public hospitals’ corporatization are also an area for future research. References Alexander, J., D’Aunno, T., & Succi, M. J. (1996). Determinants of rural hospital conversion: A model of profound organizational change. Medical Care, 34(1), 29Y43. Almwajeh, O. (2004). Applying Altman’s Z-score model of bankruptcy for the prediction of financial distress of rural hospitals in Western Pennsylvania. Doctoral dissertation, Indiana University of Pennsylvania, Indiana. Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance, 23(4), 589Y609. Altman, E. I. (1993). Corporate financial distress and bankruptcy (2nd ed.). New York, NY: John Wiley & Sons.


Altman, E. I. (2002). Revisiting credit scoring models in a Basel 2 environment. Unpublished manuscript. Retrieved from http:// citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.196.55& rep=rep1&type=pdf American Hospital Association. (2009). Teaching hospitals: Their impact on patients and the future of health care workforce. Retrieved from http://www.aha.org/research/rc/stat-studies/Studies.shtml American Hospital Association. (2012). Facts on US hospitals. Retrieved from http://www.aha.org/research/rc/stat-studies/fast-facts.shtml Amirkhanyan, A. (2007). The Smart-seller challenge: Exploring the determinants of privatizing public nursing homes. Journal of Public Administration Research and Theory, 17(3), 501Y527. Bachrach, D., Braslow, L., & Karl, A. (2012). Toward a high performance health care system for vulnerable populations: Funding for safetynet hospitals. Retrieved from http://www.commonwealthfund.org/ Publications/Fund-Reports/2012/Mar/Vulnerable-Populations.aspx Bannon, E. P. (2013). Nine of Louisiana’s ten public hospitals to be privatized. World Socialist Web Site. Retrieved from http://www .wsws.org/en/articles/2013/05/30/loui-m30.html Bazzoli, J. G., & Cleverley, W. O. (1994). Hospital bankruptcies: An exploration of potential causes and consequences. Health Care Management Review, 19(3), 41Y51. Bovbjerg, R. R., Marsteller, J. A., & Ullman, F. C. (2000). Health care for the poor and uninsured after a public hospital’s closure or conversion. Washington, DC: The Urban Institute. Retrieved from http://www.urban.org/Uploadedpdf/309647_occa39.pdf Burns, L., Shah, R. J., Sloan, F. A., & Powell, A. (2009). The impact of hospital ownership conversions: Review of the literature and results from a comparative field study. Advances in Health Care Management, 8, 171Y229. Calandro, J. J. (2007). Considering the utility of Altman’s Z-score as a strategic assessment and performance management tool. Strategy & Leadership, 35(5), 37Y43. California Association of Public Hospitals and Health Systems. (2003). On the brink: How the crisis in California’s public hospitals threatens access to care for millions. Retrieved from http://caph .org/wp-content/uploads/PDFs/papers/118.pdf Desai, K. R., Lukas, C. V., & Young, D. W. (2000). Public hospitals: Privatization and uncompensated care. Health Affairs, 19(2), 167Y172. Field, A. (2009). Discovering statistics using SPSS. Thousand Oaks, CA: SAGE Publications. Gapenski, L. C. (2005). Healthcare finance: An introduction to accounting and financial management (3rd ed.). Chicago, IL: Health Administration Press. Hall, M., & Weiss, L. (1967). Firms size and profitability. The Review of Economics and Statistics, 49(3), 319Y331. Hannan, M. T., & Freeman, J. (1984). Structural inertial and organizational change. American Sociological Review, 49(2), 149Y164. Hart, O. (1995). Firms, contracts and financial structure. Oxford, UK: Oxford University Press. Health Affairs. (2012, August 16). Graduate medical education (health policy brief). Retrieved from http://healthaffairs.org/healthpolicybriefs/ brief_pdfs/healthpolicybrief_73.pdf Kim, T. H. (2010). Factors associated with financial distress of nonprofit hospitals. The Health Care Manager, 29(1), 52Y62. Langabeer, J. (2006). Predicting financial distress in teaching hospitals. Journal of Health Care Finance, 33(2), 84Y92. Langabeer, J. (2008). Hospital turnaround strategies. Hospital Topics, 86(2), 3Y12. Legnini, M. L., Anthony, S. E., Wicks, E. K., Meyer, J. E.,


347


Rybowski, L. S., & Stepmick, L. S. (1999). Privatization of public hospitals. Oakland, CA: The Henry J. Kaiser Family Foundation. Retrieved from http://kaiserfamilyfoundation.files.wordpress. com/1998/08/1450-privatization-of-public-hospitals-report.pdf Li, B., Lingsma, H., Steyerberg, E., & Lesaffre, E. (2011). Logistic random effects regression models: A comparison of statistical packages for binary and ordinal outcomes. BMC Medical Research Methodology, 11, 1Y11. Retrieved from http://www.biomedcentral .com/content/pdf/1471-2288-11-77.pdf Mahoney, J. T. (2005). Economic foundations of strategy. Thousand Oaks, CA: Sage Publications, Inc. McCue, M., & Furst, R. (1986). Financial characteristics of hospitals purchased by investor-owned chains. Health Services Research, 21(4), 515Y526. Pennings, J. M., & Tripathi, R. (1978). The organizationY environment relationship: Dimension versus typological viewpoints. Beverly Hills, CA: Sage Studies in International Sociology. Pfeffer, J., & Salancik, G. (1978). The external control of organizations: A resource dependence perspective. New York, NY: Harper & Row. Pink, G. H., Holmes, M., D’Alpe, C., Strunnk, L. A., McGee, P., & Slifkin, R. (2005). Financial indicators for critical access hospitals: Flex Monitoring Team. University of Minnesota, University of North Carolina at Chapel Hill, University of Southern Maine. Retrieved from http://www.flexmonitoring.org/wp-content/ uploads/2013/07/BriefingPaper7_FinancialIndicators.pdf Preker, A. S., & Harding, A. (Eds.). (2003). Innovations in health services delivery: The corporatization of public hospitals. Washington, DC: The World Bank. Retrieved from http://www-wds.worldbank. org/external/default/WDSContentServer/IW3P/IB/2003/06/ 30/000094946_0306190405383/Rendered/PDF/multi0page.pdf Ramamonjiarivelo, Z., Weech-Maldonado, R., Hearld, L., & Pradhan, R. (2014). Public hospitals in peril: Factors associated with financial distress. Journal of Health Care Finance, 40(3), 14Y30. Sataline, S. (2010). Cash-poor governments ditching public hospitals. The Wall Street Journal, 30, 1Y4. Sharfman, M., Wolf, G., Chase, R., & Tansik, D. (1988). Antecedents of organizational slack. Academy of Management Review, 13(4), 601Y614.

Shen, Y.-C. (2003). Changes in hospital performance after ownership conversions. Inquiry - Excellus Health Plan, 40(3), 217Y234. Shuit, G. (1996). UC hospitals edging closer to privatization. Los Angeles Times. Retrieved from http://articles.latimes.com/ 1996-05-16/news/mn-4763_1_uc-san-diego Sloan, F. A., Ostermann, J., & Conover, C. J. (2003). Antecedents of hospital ownership conversions, mergers, and closures. Inquiry, 40(1), 39Y56. Tiemann, O., & Schreyo¨gg, J. (2009). Effects of ownership on hospital efficiency in Germany. Business Research, 2(2), 115Y145. Tiemann, O., & Schreyo¨gg, J. (2010). Investigating the effects of hospital privatization on performance and quality of care. Paper presented at the meeting of Beitra¨ge zur Jahrestagung des Vereins fu¨r Socialpolitik 2010. Retrieved from https://www.econstor.eu/ dspace/bitstream/10419/37374/1/VfS_2010_pid_417.pdf Tradewell, R. (1998). Privatizing public hospitals: Strategic options in an era of industry-wide consolidation (Policy Study No. 242). Retrieved from (Reason Foundation Website) http://reason.org/ files/f5ae49d45881eb1b195e591a1b425c51.pdf Villa, S., & Kane, N. (2013). Assessing the impact of privatizing public hospitals in three American states: Implications for universal coverage. Value in Health, 16(1), S24YS33. Vujicic, M., Addai, E., & Bosomprah, S. (2009). Measuring health workforce productivity: Application of a simple methodology in Ghana. Washington, DC: The International Bank for Reconstruction and Development/The World Bank. Weech-Maldonado, R., Laberge, A., Pradhan, R., Johnson, C. E., Yang, Z., & Hyer, K. (2012). Nursing home financial performance: The role of ownership and chain affiliation. Health Care Management Review, 37(3), 235Y245. Weech-Maldonado, R., Qaseem, A., & Mkanta, W. (2009). Operating environment and USA nursing homes’ participation in the subacute care market: A longitudinal analysis. Health Services Management Research, 22(1), 1Y7. Weil, T. P. (2011). Privatization of hospitals: Meeting divergent interests. Journal of Health Care Finance, 38(2), 1Y11. Wooldridge, J. M. (2006). Introductory econometrics: A modern approach. Mason, OH: Thomson South-Western.


Eye Movements in Strategic Choice.

Financial Distress in Cancer Patients.

Strategic action grids: a study in Indian hospitals.

Financial distress in patients with advanced cancer.

Keeping strategic thinking in strategic planning: macro-environmental analysis in a state department of public health.

Relationships Among Financial Distress, Emotional Distress, and Overall Distress in Insured Patients With Cancer.

IT strategic planning in hospitals: from theory to practice.

[Distress Among Surgeons - a Study in German Hospitals.]

The role of general hospitals in the privatization of inpatient treatment for serious mental illness.

Financial Analysis of National University Hospitals in Korea.

Herbal drugs and public choice.

Soft Budget Constraints in Public Hospitals.

[Prevalence of pain in Andalusian public hospitals].

A literature review of the regional implementation of the central Swedish government's health care reforms on choice and privatization.

Patient-oncologist cost communication, financial distress, and medication adherence.

Provision of essential health package in public hospitals: a case of Homabay County hospitals, Kenya.

A Study on Strategic Planning and Procurement of Medicals in Uganda's Regional Referral Hospitals.

Erratum to: 'A cost and performance comparison of Public Private Partnership and public hospitals in Spain'.

A cost and performance comparison of Public Private Partnership and public hospitals in Spain.

Strategic planning for public health practice using macroenvironmental analysis.

Prenatal screening, reproductive choice, and public health.

Medicalizing distress, ignoring public health strategies.

NSW public hospitals accessing digital radiology images.

Good public awareness about albinism is key to reduced psychiatric distress in African people with albinism.